Key findings
- Avoid no-name options as they lead to higher temperatures and instability.
- However, there is no need to spend extra money on high-quality thermal pastes.
- Unless you're an expert at overclocking, inexpensive (brand-name) thermal paste can provide similar stability and temperature to premium options.
Unless you're building a battle station or gaming PC with the most expensive parts money can buy, it's a good idea to spend a higher percentage of your budget on select components. For example, you should spend more than half of your PC build budget on the graphics card, with the processor earning the second highest priority.
However, thermal paste is always overlooked when estimating costs, mainly because even the highest-end options cost a fraction of the overall budget. However, if you're looking to minimize spending, you may come across some no-name options that offer large quantities of thermal paste at bargain prices. After purchasing such a kit a few months ago, I decided to compare it to the Corsair TM30 to see if it was worth the purchase. Spoiler: It's not.
Disclaimer: When I say no-name brand thermal pastes, I mean the cheapest options that contain ten or even twenty times more than comparably priced branded products. Of course, the TM30 doesn't come close to the best thermal pastes out there, but it's still good enough to be considered a high-quality thermal paste.
The test bench for this experiment
Since I wanted to test both thermal pastes under heavy load, I chose to use my Ryzen 5 5600X processor, which is liquid cooled by a Kraken X52 AIO. I also enabled PBO and Resizable BAR in the BIOS and used an RTX 3080 Ti to avoid GPU bottlenecks when running the gaming tests. I had also changed the power plan in the Control Panel to squeeze the maximum performance out of every component, including the CPU. Finally, I prevented my motherboard from hibernating the idle CPU cores by disabling the “Global C-States” option in the BIOS. Although this option pushed my idle temperatures up a bit, I wanted to see the maximum performance and temperature the CPU could reach with each thermal paste.
The stress test of the thermal paste without brand name showed disturbing Maximum temperatures …
And I was a little afraid that my CPU would burn out in the middle of the test
Speaking of testing, I used different methods to check the system's temperatures and stability with different thermal pastes. After cleaning the existing paste on my Ryzen 5 5600X, I decided to test the unbranded thermal paste first. After booting into Windows, I left the PC in standby mode for a few minutes to measure idle temperatures. Next, I ran the Intel Burn Test – and to make things even more interesting, I subjected it to the Blend torture tests on Prime95.
Since I was using PBO, I expected the system to crash at least once. But to my surprise, the cheap thermal paste got through without causing any stability issues. However, checking the temperature readings in HWiNFO64 revealed a strange trend: although the average temperatures were a bit higher than normal, the maximum temperatures reached the red zone more than a few times.
CPU state/apps |
idle |
idle |
Intel Burn Test |
Intel Burn Test |
Prime95 |
Prime95 |
---|---|---|---|---|---|---|
Paste used |
Thermal paste without brand name |
High-quality thermal paste |
Thermal paste without brand name |
High-quality thermal paste |
Thermal paste without brand name |
High-quality thermal paste |
CPU (Tctl/Tdie) |
|
|
|
|
|
|
CPU Chip (Average) |
|
|
|
|
|
|
CPU CCD1 (Tdie) |
|
|
|
|
|
|
Core temperatures |
|
|
|
|
|
|
So before I discuss the gaming trends, it's time to fast forward a few hours when I repeated these tests with the premium thermal paste. And as expected, the average and maximum temperatures were lower this time, especially in the Intel Burn test. Meanwhile, in Prime95, the premium thermal paste had only a few spikes above the 90°C range, while its unbranded competitor was already averaging 93°C.
…And the situation was not much different on the gaming front
At least the frame rates were not too different
Although stress tests are a good way to measure the stability of a processor, they tend to put the CPU under unrealistic strain, so I decided to compile the benchmarks and graphs I got after running some games. For each title, I turned the resolution down to 1080p while leaving the settings on Ultra to force my CPU to spread its wings.
Games |
Baldur’s Gate 3 |
Baldur’s Gate 3 |
Cyberpunk2077 |
Cyberpunk2077 |
Red Dead Redemption 2 |
Red Dead Redemption 2 |
---|---|---|---|---|---|---|
Paste used |
Thermal paste without brand name |
High-quality thermal paste |
Thermal paste without brand name |
High-quality thermal paste |
Thermal paste without brand name |
High-quality thermal paste |
CPU (Tctl/Tdie) |
|
|
|
|
|
|
CPU Chip (Average) |
|
|
|
|
|
|
CPU CCD1 (Tdie) |
|
|
|
|
|
|
Core temperatures |
|
|
|
|
|
|
In terms of FPS, there wasn't too much difference between titles. But again, the unbranded thermal paste had higher maximum temperatures in almost all cases, with a few exceptions. After both the stress tests and gaming benchmarks were completed, I had enough evidence to make my judgment.
Is non-branded thermal paste worth the money?
Absolutely not! If you care about the longevity of your systems, you don't have to resort to low-quality thermal paste. Luckily, there are plenty of affordable alternatives that will keep your processor nice and cool. In fact, you won't notice much difference in terms of stability and temperatures if you choose a budget option over a high-quality thermal paste. Just avoid no-name brands, and as long as your gaming machine has adequate cooling provisions, you should be fine.